The extrusion processing of instant coffee for the purposes of creating a pressurised microstructure

Abstract

This thesis is concerned with the derivation, development, and validation of a model for the dynamic behaviour of gas bubbles in molten instant coffee, in response to changes in pressure and temperature.

Initially, the material properties of molten instant coffee were investigated. The primary technique used was capillary rheometry, as the viscous behaviour of molten instant coffee is a major contributor to the behaviour of bubbles within the fluid. The experimental technique was supplemented by analytical and computational modelling of the pressure losses in the system, to improve data processing accuracy. Additional thermal and mechanical tests were performed to obtain as much of the necessary material information as possible for the system.

The modelling of bubble behaviour was performed using a combination of analytical and computational methods. The relationship between the pressure driving force for bubble growth or shrinkage and the rate of change of bubble size was derived analytically for a number of common generalised Newtonian fluid models. Heat and mass transfer between the bubble and the surrounding fluid was calculated using a finite difference approximation of the governing partial differential equations. The model was written in MATLAB and initial validation was carried out by comparison with existing models for bubble dynamics.

Experimental observations of bubble dynamics in flows of molten instant coffee were recorded and used for an extended validation of the model. Bubbles were exposed to step changes in pressure and oscillatory pressure profiles at a range of temperatures, and the observations of 130 individual bubbles were used to validate the model using the same material parameters for each. A final case study in using the bubble model to predict the bubble size and pressure distribution created by an extrusion process is presented as an example of the use of the model, and highlights the additional information about a process that is required to effectively use the model.